---
title: "How to perform CCI simulation by `cellCellSimulate` function"
author:
- name: Koki Tsuyuzaki
  affiliation: Laboratory for Bioinformatics Research,
    RIKEN Center for Biosystems Dynamics Reseach
- name: Manabu Ishii
  affiliation: Laboratory for Bioinformatics Research,
    RIKEN Center for Biosystems Dynamics Reseach
- name: Itoshi Nikaido
  affiliation: Laboratory for Bioinformatics Research,
    RIKEN Center for Biosystems Dynamics Reseach
  email: k.t.the-answer@hotmail.co.jp
package: scTensor
output:
  BiocStyle::html_document
vignette: |
  %\VignetteIndexEntry{scTensor: 3. Simulation of CCI}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

# Introduction

Here, we explain the way to generate CCI simulation data.
`r Biocpkg("scTensor")` has a function `cellCellSimulate`
to generate the simulation data.

The simplest way to generate such data is `cellCellSimulate` with default parameters.

```{r cellCellSimulate_Default, echo=TRUE}
suppressPackageStartupMessages(library("scTensor"))
sim <- cellCellSimulate()
```

This function internally generate the parameter sets by `newCCSParams`,
and the values of the parameter can be changed, and specified as the input of `cellCellSimulate` by users as follows.

```{r cellCellSimulate_Setting, echo=TRUE}
# Default parameters
params <- newCCSParams()
str(params)

# Setting different parameters
# No. of genes : 1000
setParam(params, "nGene") <- 1000
# 3 cell types, 20 cells in each cell type
setParam(params, "nCell") <- c(20, 20, 20)
# Setting for Ligand-Receptor pair list
setParam(params, "cciInfo") <- list(
    nPair=500, # Total number of L-R pairs
    # 1st CCI
    CCI1=list(
        LPattern=c(1,0,0), # Only 1st cell type has this pattern
        RPattern=c(0,1,0), # Only 2nd cell type has this pattern
        nGene=50, # 50 pairs are generated as CCI1
        fc="E10"), # Degree of differential expression (Fold Change)
    # 2nd CCI
    CCI2=list(
        LPattern=c(0,1,0),
        RPattern=c(0,0,1),
        nGene=30,
        fc="E100")
    )
# Degree of Dropout
setParam(params, "lambda") <- 10
# Random number seed
setParam(params, "seed") <- 123

# Simulation data
sim <- cellCellSimulate(params)
```

The output object **sim** has some attributes as follows.

Firstly, **sim$input** contains a synthetic gene expression matrix.
The size can be changed by **nGene** and **nCell** parameters described above.

```{r input, echo=TRUE}
dim(sim$input)
sim$input[1:2,1:3]
```

Next, **sim$LR** contains a ligand-receptor (L-R) pair list.
The size can be changed by **nPair** parameter of **cciInfo**,
and the differentially expressed (DE) L-R pairs
are saved in the upper side of this matrix.
Here, two DE L-R patterns are specified as **cciInfo**,
and each number of pairs is 50 and 30, respectively.

```{r LR, echo=TRUE}
dim(sim$LR)
sim$LR[1:10,]
sim$LR[46:55,]
sim$LR[491:500,]
```

Finally, **sim$celltypes** contains a cell type vector.
Since **nCell** is specified as "c(20, 20, 20)" described above,
three cell types are generated.

```{r celltypes, echo=TRUE}
length(sim$celltypes)
head(sim$celltypes)
table(names(sim$celltypes))
```

# Session information {.unnumbered}

```{r sessionInfo, echo=FALSE}
sessionInfo()
```